Analysis of Nd3+ concentration on the structure, morphology and photoluminescence of sol-gel Sr3ZnAl2O7 nanophosphor

Accepted Manuscript Analysis of Nd3+ concentration on the structure, morphology and photoluminescence of sol-gel Sr3ZnAl2O7 nanophosphor M.R. Mhlongo, L.F. Koao, T.E Motaung, R.E. Kroon, S.V. Motloung PII: DOI: Reference:

S2211-3797(18)31027-1 https://doi.org/10.1016/j.rinp.2019.02.004 RINP 2068

To appear in:

Results in Physics

Received Date: Revised Date: Accepted Date:

14 May 2018 1 February 2019 1 February 2019

Please cite this article as: Mhlongo, M.R., Koao, L.F., Motaung, T.E, Kroon, R.E., Motloung, S.V., Analysis of Nd3+ concentration on the structure, morphology and photoluminescence of sol-gel Sr3ZnAl2O7 nanophosphor, Results in Physics (2019), doi: https://doi.org/10.1016/j.rinp.2019.02.004

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Analysis of Nd3+ concentration on the structure, morphology and photoluminescence of sol-gel Sr3ZnAl2O7 nanophosphor

M.R. Mhlongoa,*, L.F. Koaob, T.E Motaungc, R.E. Kroond, S.V. Motlounge a

Department of Physics, Sefako Makgatho Health Sciences University, P.O. Box 94, Medunsa, 0204, South Africa

b

Department of Physics, University of the Free State (Qwaqwa Campus), Private Bag X 13, Phuthaditjhaba, 9866, South Africa c

Department of Chemistry, University of Zululand, KwaDlangezwa, 3886, South Africa

d

Department of Physics, University of the Free State, P. O. Box 339, Bloemfontein, 9300, South Africa

e

Department of Physics, Nelson Mandela University (NMU), P. O. Box 77000, Port Elizabeth 6031, South Africa

Corresponding author: [email protected]

Abstract Neodymium activated strontium zinc aluminate (Sr3ZnAl2O7:x%Nd3+) nanophosphor was synthesized using the sol-gel technique whereby the Nd3+ concentration was varied in the range 0 ≤ x ≤ 2. The effect of Nd3+ concentration on the structure, particle morphology and photoluminescence properties of Sr3ZnAl2O7 were investigated. The X-ray diffraction (XRD) results revealed that all samples resembled the mixture of both ZnAl2O4 and Sr3Al2O6 cubic structures. Nd3+ doping influenced the crystallite sizes of the prepared phosphor materials. The energy dispersive X-ray spectroscopy (EDS) results confirmed the presence of all expected elements in the composition. Scanning electron microscopy (SEM) revealed that as the Nd3+ concentration increased the surface morphology changed to smooth mountain-like structures. The ultraviolet-visible (UV-Vis) diffuse reflection spectroscopy showed that the band gap of Sr3ZnAl2O7 can be tuned from 2.74 to 2.95 eV by increasing the Nd3+ concentration. When the host is excited above the bandgap (374 nm), broad emission attributed to defects occurs with the maximum near 585 nm. Doped samples excited in this manner do not exhibit additional luminescence due to the Nd3+ ions, but in contrast there is a small dip in the defect emission band near 585 nm due to absorption attributed to Nd3+ ions. Characteristic infrared emissions of Nd3+ ions at 885, 1064 and 1340 nm were observed by directly exciting the Nd3+ ions at 585 nm (4I9/2 → G5/2+2G7/2) and were attributed to 4F3/2 → 4I9/2, 4I11/2 and 4I13/2 transitions, respectively. The

5

Commission Internationale de l’Eclairage (CIE) coordinates results showed that the orange emission colour is from the host. 1

Keywords: Sol-gel, Nd3+ activated, Sr3ZnAl2O7, photoluminescence, CIE.

1. Introduction In recent years the growing demand for luminescence materials for various applications has triggered targeted studies to explore new phosphor materials. Aluminate phosphors have been extensively investigated and have been used for photoluminescence (PL) and plasma display panels because of their high efficiency, wide excitation wavelength and high quenching temperature [1]. Among the aluminates, strontium aluminate (SrAl2O4) and zinc aluminate (ZnAl2O4) spinels are considered as ideal host materials for long-lasting phosphorescence [1]. SrAl2O4 changes structure to Sr3Al2O6 at around 800 oC. Sr3Al2O6 with an energy band gap (Eg) of 6.3 eV has a cubic structure with space group Pa-3 (No. 205) [2,3]. On the other hand, ZnAl2O4 has an Eg of 3.8 eV with a face-centred cubic structure with Fd3m space group symmetry. ZnAl2O4 is known to have high mechanical resistance, high fluorescence efficiency, high chemical and thermal stability, high photocatalytic activity, and low surface acidity [4-6]. Recently, ZnAl2O4 has been studied as a phosphor host material for application in thin film electroluminescence displays, mechano-optical stress sensors and stress imaging devices [7]. ZnAl2O4 and SrAl2O4 have the normal spinel structure and the chemical formula of AB2O4 [A = Zn or Sr] in which Zn2+ or Sr2+ (A) ions occupy tetrahedral sites and Al3+ (B) ions octahedral sites. Many researchers have explored several methods such as sol-gel [8], combustion [9], solid-state reaction [10], and co-precipitation [11], electrospinning [12] or floating zone (FZ) [13] to successfully fabricate phosphor materials. In comparison with other methods, the sol-gel method gives several advantages such as good stoichiometric control, high homogeneity, convenience, cost effectiveness, small and uniform particle size at relatively low temperature, and the ability to produce nanostructured powders and thin films [14,15]. Thus the sol-gel method was employed in this study. SrAl2O4 and ZnAl2O4 are efficient host materials with a wide Eg that generates broadband emission upon doping with rare-earth (RE) ions or transition metal (TM) ions [16,17]. Nd3+ is one of the frequently reported RE ions which leads to NIR luminescence and has significant applications in biological and optical fields. Ayvacıkli et al. [18] synthesized SrAl2O4:Mn2+, Nd3+ by solid-state reaction method and the absorption lines for upper energy states were observed at 431, 475, 514, 527, 583, 626, 682 and 748 nm from the visible to NIR wavelengths. These transitions were attributed to the transitions from the ground level of 4I9/2 to the higher levels of Nd3+. Also the three emission bands at 878, 1059 and 1333 nm when excited by 473 nm were assigned to the 4F3/2 → 4I9/2, 4

I11/2 and 4I13/2 transitions of Nd3+, respectively. Similar results were observed by Yang et al. [12] on

the study of ZnAl2O4:Nd3+ using the electrospinning technique. The PL results revealed that under 2

808 nm excitation wavelength, there were three emission peaks at 905, 1064 and 1335 nm, which were respectively attributed to the same transitions mentioned previously. The optimal doping concentration of Nd3+ was found to be 1% . Keeping in mind the advantages of Sr3Al2O6 and ZnAl2O4 discussed above and the influence that Nd

3+

ions have on these host materials, in this study the idea is to develop or synthesize the

Sr3ZnAl2O7 (host) material which may have different and beneficial properties compared to the bulk Sr3Al2O6 and ZnAl2O4. Reports on a mixture of these two host (Sr3Al2O6 and ZnAl2O4) materials such as Sr3ZnAl2O7:x% Nd3+ are very scarce. Therefore, in this study, we are interested to investigate and explore the material properties of Sr3ZnAl2O7:x% Nd3+. The effect of Nd3+ concentration on the structural, morphological and photoluminescence properties of Sr3ZnAl2O7:x% Nd3+ are discussed in detail with the main aim of producing an alternative nanophosphor material, which could be used for practical applications such as in biological and optical fields. The physics behind the emission pathways or mechanisms is also proposed and explained.

2. Experimental 2.1 Powder sample synthesis Nanophosphors were prepared using the citrate sol-gel technique [14]. Undoped Sr3Al2O6, ZnAl2O4, Sr3ZnAl2O7 were prepared by putting Sr(NO3)2 (98%), Zn(NO3)2·6H2O (98%) and Sr(NO3)2 (98%) + Zn(NO3)2·6H2O (98%) into separate beakers. Al(NO3)3·9H2O (98%) and citric acid (CA) C8H8O7·H2O (99%) were added in those beakers and the mixtures were dissolved in deionised water. The doped samples Sr3ZnAl2O7:x% Nd3+ were prepared by dissolving Sr(NO3)2 (98%), Zn(NO3)2·6H2O (98%), Al(NO3)3·9H2O (98%), C8H8O7·H2O (99%) and Nd(NO3)3·6H2O (99%) in deionised water. Nd(NO3)3·6H2O (99%) was added at the range (0.6% ≤ x ≤ 2%). The sols stoichiometric molar ratio of Sr:Zn:Al and Sr:Zn:CA were 1:1:2 and 1:1:0.075, respectively. A magnetic stirrer was used to stir the solution at a constant temperature of 80 oC until a gel was formed. With increasing Nd3+ concentration the gel became milky. The gel was left to dry at room temperature for 2 h and then annealed in a furnace at 1000 oC for 2 h. The solid product from the furnace was crushed using a mortar and pestle in order to prepare the powder samples which were then analysed using different techniques.

2.2 Samples characterization

The crystal structure and phase purity of the samples were characterized by powder X-ray diffraction (XRD) (Bruker AXS Discover diffractometer) with Cu Kα (1.5418A) radiation. A Shimadzu Superscan Zu SSX-550 electron microscope (SEM) coupled with the energy dispersion spectroscopy (EDS) was used to investigate the morphology and element composition. Ultraviolet3

visible (UV-Vis) diffuse reflection spectroscopy was used to study the absorption characteristics of the prepared samples. Luminescence measurements were made with an Edinburgh Instruments FLS980 fluorescence spectrometer having double monochromators, using a steady state xenon lamp as excitation source and R928P and R5509-72 photomultiplier tubes for visible and infrared emissions, respectively. Lifetime measurements were obtained using a Hitachi F-7000 fluorescence spectrophotometer. All characterizations were performed at room temperature.

3. Results and discussion 3.1 XRD analysis Fig. 1 shows the XRD patterns of the un-doped Sr3Al2O6, ZnAl2O4, Sr3ZnAl2O7 (host) and Sr3ZnAl2O7:x% Nd3+ powder samples, respectively. Fig. 1 (a) illustrates that all of the diffraction peaks for the ZnAl2O4 and Sr3Al2O6 respectively matched very well with the face-centered cubic ZnAl2O4 spinel (JCDPS card 82-1043) and primitive cubic Sr3Al2O6 (JCDPS card 81-0506). It can also be confirmed in Fig. 1 (a) that the diffraction patterns for the calculated Sr3ZnAl2O7 resembles the mixture of both ZnAl2O4 and Sr3Al2O6 which suggest that the host material is a mixture of these. The Sr3ZnAl2O7:x% Nd3+ samples shown in Fig. 1(b) exhibited similar diffraction patterns as the host without any impurities irrespective of the increase in Nd3+ concentration. This implies that the Nd3+ ions were successfully incorporated into the host crystal lattice without altering the host material structure. The lattice parameters were calculated from equation 1 [9]

(1)

where a is the lattice parameter,

is the interplanar distance and

are the Miller indices. The

lattice parameters for both ZnAl2O4 and Sr3Al2O6 were calculated from peaks (220) and (440) to be 8.013 and 15.770 Å, respectively. These values are comparable to a = 8.080 Å (ZnAl2O4) [19] and a = 15.850 Å (Sr3Al2O6) [20], which were previously reported in literature. The incorporated Nd3+ in the crystal structure of the host might have been by substitution on either Sr 2+, Zn2+ or Al3+ sites [21]. Fig. 2 shows the analysis of the diffraction peaks (220) around 30o from the ZnAl2O4 and (440) around 32o from the Sr3Al2O6 in the mixed materials. Compared to the host sample, the increase of the (440) peak intensity as the Nd3+ increases can be associated to the increase of the crystal quality of the host, whereas the decrease in (440) and (220) peaks is associated with the deterioration of the crystalline quality. Thus, varying the Nd3+ concentration influences the crystallinity of the host sample. It can be observed that for the lower Nd3+ concentration (x ≤ 1) there is a shift to the lower angles indicating that the lattice parameters are larger than that of the host [22]. The shifting of the peaks to the lower angles can be attributed to the larger dopant Nd3+ (1.16 Å) substituting the smaller Al3+ (0.53 Å) [18, 4

22] or Zn2+ (0.74 Å) [23] ion in the host crystal lattice [24,25]. However, when the Nd3+ concentration is 1.4 ≤ x ≤ 2 it is observed that the diffraction angle shifts to the higher angles suggesting the decrease on the lattice parameters [26]. This is furthermore confirmed by the calculated values shown in Table 1. The shift to the higher angles can be explained by assuming that Nd3+ (1.16 Å) [18] replace the larger Sr2+ (1.21 Å) [18] on the host crystal lattice. The lattice parameter as a function of Nd3+ concentration for the Sr3ZnAl2O7:x% Nd3+ (0 ≤ x ≤ 2) series is presented in Fig. 3 (a) and (b). The crystallite sizes (D) of the Sr3ZnAl2O7 (0 ≤ x ≤ 2) series were estimated using the Scherrer equation [27]:

D

0.9  cos 

(2)

where λ is the wavelength of the incident X-rays, θ is the diffraction angle and β is the full width of the diffraction line at half maximum intensity (FWHM), in radians. The estimated average crystallite sizes are presented in Table 1. Crystallite sizes as a function of Nd3+ concentration are illustrated in Fig. 3 (c) and (d), which show the Gaussian behaviour. This behaviour can be correlated to the broadening of the XRD peaks due to doping on the host material [28].

5

Fig. 1. The XRD pattern for the (a) Sr3Al2O6, ZnAl2O4 and Sr3ZnAl2O7 undoped samples and (b) Sr3ZnAl2O7:x% Nd3+ series.

6

Fig. 2. Analysis of the ZnAl2O4 (220) and Sr3Al2O6 (440) diffraction peaks In order to evaluate the strain caused by incorporation of the Nd3+ into the host crystal lattice, the Williamson-Hall method [29] was employed. The calculated values for each sample for the Sr3ZnAl2O7 (0 ≤ x ≤ 2) series are presented in Table 1. Fig. 3 (e) and (f) shows the strain as a function of Nd3+ concentration for the diffraction peaks (220) and (440), respectively. Generally, the strain increases with an increase in Nd3+ doping concentration, which can be attributed to the incorporation of more foreign ions (Nd3+) into the crystal structure of Sr3ZnAl2O7 as the Nd3+ concentration is increased. The incorporation of foreign ions or atoms into the crystal structure of the host material is expected to cause the defects when the dopant Nd3+ replaces either Zn2+ Al3+ or Sr2+ [30]. These defects are due to the substitutions of different sizes of the atoms or ions as discussed in Fig. 2. Fig 3 (e) and (f) also shows the dislocation density as a function of Nd3+ concentration, which resembles similar-kind of behaviour as the strain and this is attributed to the increase of more substitutions of atoms as the Nd3+ concentration is increased. Motloung et al. [30] obtained a similar trend in ZnAl2O4:0.1% Ce3+, x% Eu3+ system.

7

Table 1. Summary of the samples identification, lattice parameter, crystallites size, strain, and dislocation density Sample ID

Lattice parameter

Crystallite size

Strain

Dislocation Density x10-3

a (Å)

(nm)

(nm-2)

(%)

(220)

(440)

(220)

(440)

(220)

(440)

(220)

(440)

Sr3Al2O6

-

15.776

-

62.34

-

0.73

-

0.26

ZnAl2O4

8.013

-

23.13

-

2.02

-

1.86

-

Host

8.037

15.783

35.98

35.99

1.30

1.28

0.77

0.77

x = 0.6%

8.047

15.799

36.39

36.57

1.29

1.26

0.76

0.75

x = 0.8%

8.047

15.805

34.85

36.08

1.35

1.27

0.82

0.77

x = 1%

8.041

15.794

34.23

29.66

1.37

1.54

0.85

1.14

x = 1.4%

8.021

15.754

25.21

28.33

1.86

1.62

1.57

1.24

x = 1.8%

8.019

15.737

22.44

25.74

2.09

1.78

1.99

1.51

x = 2%

8.024

15.748

21.78

25.47

2.15

1.80

2.11

1.54

8

Fig. 3 (a) Lattice parameter as a function of Nd3+ (220) (b) Lattice parameter as a function of Nd3+ (440) (c) crystallite sizes as a function of Nd3+ (220) (d) crystallite sizes as a function of Nd3+ (440) (e) strain and dislocation density as a function of Nd3+ (220) (f) strain and dislocation density as a function of Nd3+ (440).

9

3.2 EDS analysis The EDS spectra of the host and Sr3ZnAl2O7:2% Nd3+ are illustrated in Fig. 4. Fig 4 (a) confirms the existence of the Sr, Zn, Al and O on the host material as expected. In Fig. 4 (b) there are additional peaks of Nd which confirms that the host material was doped with Nd3+. The samples were carbon coated and mounted on the carbon tape during EDS measurements and hence there are carbon (C) peaks present in both spectra. The EDS and the XRD results in Fig. 1 (b) are in agreement because only expected elements are present together with C. The EDS was further used to confirm the elemental distribution on the surface of both samples. Fig. 5 show the EDS elemental maps of the host and the Sr3ZnAl2O7:2% Nd3+. It can be seen that all of the anticipated elements are distributed homogeneously on the surface.

Fig. 4. EDS spectrum of the (a) host and (b) Sr3ZnAl2O7:2%Nd3+.

10

Fig. 5. Elemental map of (a) the host (b) Sr3ZnAl2O7:2%Nd3+.

11

3.3 SEM results The morphological feature of the prepared powders was explored by using SEM as shown in Fig. 6. Fig. 6 (a) shows the micrograph of the Sr3Al2O6, consisting of tiny crystallites clustered together on top of bigger crystallites. These bigger crystallites are of irregular shaped agglomerates with grain boundaries. The ZnAl2O4 sample is presented in Fig. 6 (b), which shows closely packed and evenly distributed crystallites over the surface. The micrograph in Fig. 6 (c) displays the host sample which resembled the combined morphology of Sr3Al2O6 and ZnAl2O4, which agrees very well with the XRD results. At the lower Nd3+ concentration (0.6% and 1% Nd3+) the phosphor morphology resembles the one for the host material (see Fig. 6 (d) and (e)). At the high concentration 2% Nd3+ as shown in Fig. 6 (f), the morphology clearly changed to the mountain like-structures distributed over the rough surface. The results therefore suggest that the increase in Nd3+ concentration changes or influences the surface morphology of the prepared phosphor.

Fig. 6. SEM images for the (a) Sr3Al2O6 (b) ZnAl2O4 (c) host (d) 0.6% Nd3+ (e) 1% Nd3+ (f) 2% Nd3+. 12

3.4 UV–Vis UV–Vis diffuse reflection spectroscopy was used to study the absorption characteristics of the prepared samples. Fig. 7 (a) revealed that there are several absorption bands at 374, 478, 510, 528, 585, 681 and 748 nm. The absorption band at 374 nm is attributed to the band-to-band transition. The other bands at 478, 510, 528, 585, 681 and 748 nm are attributed to the f-f transitions from the ground state (4I9/2) to the higher excited states (2G9/2+2K15/2; 4G9/2; 4G7/2+4K13/2; 4G7/2+4G5/2; 4F9/2 and 4

F7/2+4S3/2), respectively [31].

Fig. 7. (a) The diffuse reflectance spectra of the host and Sr3ZnAl2O7:x% Nd3+ (0.6 ≤ x ≤ 2%) (b) Estimate of the direct optical bandgap of the samples in (a) using Kubelka-Munk function (c) Energy band gap value as a function of Nd3+ concentration

13

The Kubelka-Munk function K = (1 – R)2/2R [32] was used to transform the reflectance to the values proportional to the absorbance and the Tauc plot of (K  h  )n against h  is presented in Fig. 7 (b) where h  is the incident photon energy, n is the exponent that determines the type of electronic transition causing the absorption which can take the values ½ or 2 depending whether the transition is indirect or direct, respectively. ZnAl2O4 and Sr3Al2O6 are known to have a direct band gap because the lowest energy of the conduction band occurs at the same value of the wave vector K as the highest energy of the valence band [33] and they only require photon energy for the electron to be excited from the valence band to the conduction band. The Eg is estimated by extrapolation of the linear region of this plot to (K x hν)2 = 0. The results indicated that the Eg of the host depends on the Nd3+ concentration as shown in Fig. 7(c). It shows that the Eg increases with the doping concentration. The Eg can be tuned between 2.74 - 2.94 eV. The increase in Eg is attributed to Burstein-Moss effect [34]. The results show that when the concentration of the dopant increased, the electrons populate states within the conduction band. It pushes the Fermi level inside the conduction band towards the higher energy level, which will lead to the expansion of the energy gap. Similar type of phenomena have been observed [35,36]. Taking the XRD results into consideration, it can be noticed that the Eg increases, while the crystallite sizes decrease when the Nd3+ concentration is increased [37].

3.5 PL results Fig. 8 (a) presents the PL spectra for the excitation and emission of the Sr3ZnAl2O7:x%Nd3+ (0 ≤ x ≤ 2%) series on the visible range. Fig. 8 (a) shows that the doped samples have a sharp UV excitation peak located at 374 nm when monitoring the orange emission at 585 nm. As anticipated from the UV-vis results, the excitation at 374 nm is attributed to band-to-band transitions. When monitoring the excitation wavelength at 374 nm, the results showed that there is an emission band at 585 nm with some small shoulders at 461 and 442 nm (Fig. 8 (b)). Liu et al [38] and Jadwisienczak et al [39] reported a broad emission peak centred at 585 and 600 nm under UV excitation of 370 nm. They attributed the emissions to the defects states such as the interstitial oxygen (Oi). It can also be observed that as the Nd3+ concentration was added the dip occurs at 585 nm whereby two peaks emerge at 565 and 604 nm. It can be argued that they are not emission peaks but the edges that occurred because of the absorption by the Nd3+ ions at 585 nm. It is quite interesting to note that the emission at 585 nm from the host material corresponds to the I9/2 → H11/2 absorption band of Nd3+ as shown in UV-vis results in Fig. 7(a). This observation serves as the fundamental indication of the existence of the energy transfer (ET) from the host → Nd3+. The 585 nm wavelength was then used as the excitation wavelength to study emission in the NIR region as shown in Fig. 8(c). When monitoring the emission peak at 1064 nm, there are excitation peaks located at 429, 473, 529, 585, 680, 742 and 803 nm, which are attributed to the transitions from the ground state (4I9/2) to the higher excited states (2P1/2+2D5/2, 4G11/2+2K15/2, 4G7/2+2K13/2, 2G7/2+4G5/2, 4

F9/2, 4F7/2+ 4S3/2, 4F5/2+2H9/2), respectively [31]. When using the excitation wavelength of 585 nm, 14

Nd3+ ions in the ground state absorb the energy and get excited to the 2G7/2+4G5/2 level followed by various non-radiative (NR) relaxations. Then 4F3/2 level will be populated to produce the radiative NIR emissions at 885, 1064 and 1340 nm [18]. These emission peaks are respectively ascribed to the characteristics of 4F3/2→4I9/2, 4I1/2 and 4I13/2 transitions of Nd3+. Similar results were observed in SrAl2O4:Mn2+:Nd3+ and in Ca2BO3Cl:Eu2+:Nd3+ [18,40]. Fig. 8 (d) shows the emission intensity (of 1064 nm) as a function of Nd3+ concentration. Note that the host is excluded from the data. The results clearly indicate that when Nd3+ concentration is increased, the luminescence intensity increases and reach a maximal value and then it starts to decrease when more concentration is added. The increase and decrease in the luminescence intensity as the Nd3+ concentration increases is attributed to the luminescence enhancement and quenching [41], respectively. Luminescence enhancement might be caused by the energy transfer from the host to Nd3+. The results indicate that 0.77% Nd3+ is the optimum doping concentration for the emission at 1064 nm. The phenomena of luminescence enhancement and quenching was also observed by [24] when they increase the contents of Mgx in the study: MgxAl2O3+x:0.88% Cd2+ (0.25 ≤ x ≤ 4.5). The proposed excitation and emission pathways mechanism observed on the PL results in Fig. 8 is shown in Fig. 9. Based on the UV-vis and PL results, it is therefore reasonable to conclude that the effective Eg of the undoped (Sr3ZnAl2O7) is 374 nm (~ 3.31 eV). The radiative ET from the host material to the Nd3+ is also indicated on the emission pathways.

15

Fig. 8 (a) Emission and excitation spectra of the undoped and Sr3ZnAl2O7:Nd3+at various concentration in the visible range (b) Emission spectrum of undoped Sr3ZnAl2O7 (c) Emission and excitation spectra of the undoped and Sr3ZnAl2O7:Nd3+at various concentration in the NIR range (d) Emission intensity as a function of Nd3+.

16

Fig. 9 Emission pathways mechanism of Sr3ZnAl2O7:Nd3+ (DS is a defect state) Fig. 10 shows the PL decay curves of the prepared nanophosphors. The lifetime decay curves of the 585 nm emission (monitored at 374 nm excitation) for the prepared nanophosphor samples shown in Fig. 10 (a) were measured using a xenon flashlamp pulsing at 100 Hz together with the instrument response function (IRF). The lifetime data was fitted from the Fluoracle software of Edinburgh Instruments to extract the lifetime components and their relative intensity percentages which are presented in Table 2. The results show fast multi-exponential decays with three lifetime components (τ1, τ2 and τ3) of the order of ~2 μs, ~10 μs and ~50 μs, respectively. The IRF has only two exponential components, with lifetimes of 1.1 and 7.8 μs. The shorter lifetimes (τ 1) therefore cannot be considered as reliable values for the material, while the middle lifetimes (τ 2) may have been influenced by the instrument response. The longer lifetimes (τ3) are significantly longer than the components of the IRF and represent accurate components of the lifetime from emissions of the 17

samples. The host and the doped samples have similar decay curves, which means for this excitation wavelength (374 nm) the luminescence originated from the host material rather than the dopant. The middle life time τ2 ranging from (9.7 – 12.2 µs) was dominant with the relative intensity of around (50 – 60%). The long lifetime τ3 of the doped samples with the concentration (1.4 ≤ x ≤ 2) is longer than those with concentration (0 < x ≤ 1) with relative intensity of (20 – 22%). The host therefore produces luminescence with short lifetime limited to several tens of microseconds. As highlighted earlier in the PL discussion, an excitation wavelength of 374 nm could not excite Nd3+ ions, hence 585 nm was used. The 585 nm excitation wavelength was used to measure the lifetime of the Nd3+ 4F3/2→ 4I1/2 transition at 1064 nm. The decay curves of the 1064 nm emission are shown in Fig. 10 (b). The host material is not included since the 585 nm wavelength excite Nd 3+ ions only. The decay curves could be fitted with double exponential functions with short lifetimes τ 1 (39 80 µs) and long lifetimes τ2 (385 - 427 µs), which are both long enough that effects of the instrument response are negligible. These lifetime values suggest Nd3+ ions in two sites or two host materials, one with a longer lifetime and one where the lifetime is much less, due to greater non-radiative decay mechanisms being available.

Caponetti et al. obtained the Nd3+ lifetime value of 75 µs when

preparing YAG:Nd using conventional coprecipitation method [42].

Fig. 10. The decay curves for the emission at (a) 585 and (b) 1064 nm.

18

Table 2. Summary of the sample identification and decay times for 585 and 1064 nm emissions Sample ID

Host

Decay times (µs) for 585 nm τ1

τ2

τ3

Decay times (µs) for 1064 nm τmean

τ1

τ2

τmean

2.3 (28%)

11.1 (59%)

47 (13%)

13.5

-

-

-

3+

2.2 (33%)

9.7 (61%)

38 (6%)

9.1

68 (12%)

417 (88%)

376

0.8% Nd3+

2.2 (30%)

9.9 (62%)

39 (8%)

9.8

69 (12%)

414 (88%)

374

1% Nd3+

0.6% Nd

2.4 (29%)

10.9 (61%)

45 (10%)

11.6

80 (10%)

427 (90%)

391

1.4% Nd

3+

2.1 (22%)

12.0 (56%)

61 (22%)

20.7

62 (11%)

419 (89%)

380

1.8% Nd

3+

2.0 (23%)

11.8 (57%)

60 (20%)

19.3

39 (12%)

385 (88%)

344

2% Nd3+

2.1 (22%)

12.2 (56%)

63 (22%)

21.3

49 (9%)

408 (91%)

375

IRF

1.1 (60%)

7.8 (40%)

-

3.8

-

-

-

One of the important factors for evaluating phosphor performance is the colour coordinates. The International Commission on Illumination (CIE) chromaticity coordinates diagram for the Sr3ZnAl2O7:x%Nd3+ (0.6 ≤ x ≤ 2) series is illustrated in Fig. 11. The corresponding chromaticity coordinates (x,y) for the visible range are presented in Table 3. The results show that as the Nd 3+ concentration is increasing the emission colour does not change, this shows that all the emission in this range is from the host material. This result confirms that Nd3+ does not emit in the visible range and it is in agreement with Fig. 8 (a).

Fig. 11. Chromaticity coordinates for the host and Sr3ZnAl2O7:x%Nd3+ (0.6 ≤ x ≤ 2)

19

Table 3. Summary of the samples identification and CIE color coordinates Sample

CIE (x,y)

Host

0.150 ; 0.410

x = 0.6%

0.515 ; 0.407

x = 0.8%

0.515 ; 0.406

x = 1%

0.517 ; 0.410

x = 1.4%

0.515 ; 0.409

x = 1.8%

0.516 ; 0.406

x = 2%

0.516 ; 0.404

4. Conclusion The Sr3ZnAl2O7:x%Nd3+ (0 ≤ x ≤ 2) nanophosphors were successfully prepared using sol-gel technique. The XRD showed that the host material consist of the mixture of both the cubic structures of Sr3Al2O6 and ZnAl2O4. Variation of Nd3+ concentration influenced the lattice parameters, strain, dislocation density and crystallite sizes of the prepared nanophosphors. The EDS results confirmed the presence of the anticipated elementary composition. The SEM results showed that the morphology of the prepared samples depends on the Nd3+ concentration. UV–vis results showed that the Eg of the nanophosphor can be tuned by varying the Nd3+ concentration. The PL results demonstrated that the 585 nm emission was from the host, whereas the NIR emissions were from Nd3+. It shows that the 585 nm emission from the host corresponds to an absorption band from Nd3+, which indicates the radiative energy transfer from the host to Nd3+. The decay curves confirm the non-radiative decay mechanism.

Acknowledgement

This work is supported by the South African National Research Foundation (NRF) Thuthuka programme (fund number: UID 99266 and 113947), Sefako Makgatho Health Science University (SMU) Research Development Grant (RDG) and SMU Electron Microscope Unit. This work is based on the research supported in part by the National Research Foundation of South Africa (R.E. Kroon, Grant Number 93214). The author acknowledges Prof M Diale for her contribution towards this article.

References [1] Geng B.Y, Ma J.Z, Zhan F.M. J. Alloy Comp. 2009;473:530-533.

20

[2] Akiyama M, Xu C.N, Taira M, Nonaka K, Watanabe T. Phil. Mag. Lett. 1999;79:735. [3] Misevičius M, Jørgensen J, Kareiva A. Mat. Sci. 2013;19:438-442. [4] Duan X.L, Yuan D.R, Yu F.P. Inorg. Chem. 2011;50:5460–5467. [5] Ianos R, Borcǎnescu S, Lazaǎu R. Chem. Eng. J. 2014;240:260–263. [6] Foletto E.L, Battiston S, Simoes J.M, Bassaco M.M, Pereira L.S.F, de Moraes Flores E.M, Muller E.I. Micropor. Mesopor. Mat. 2012;163:29-33. [7] Lou Z, Hao J, Appl. Phys. A. 2005;80:151-154. [8] Motloung S.V, Dejene F.B, Swart H.C, Ntwaeaborwa O.M. J. Lumin. 2015;163:8-16. [9] Ali A.A, Karasu B, Allazov M.R, Ilyasli T.M. Chem. J. 2013;03:133-138. [10] Carter R.E. J. Am. Ceram. Soc. 1961:3:116–120. [11] Alvar E.N, Rezaei M, Alvar H.N. Synthesis, Powder. Tech. 2010;2:275–278. [12] Yang D, Zhao G, Pan Q, Liang M, Ma Z, Dong G, Dongdan C, Qiu J. Mater. Express. 2013;3: 210-216. [13] Nakauchi D, Okada G, Koshimuzi M, Yanagida T. J. Rare Earth. 2016;34:757-762.

[14] Motloung SV, Tsega M, Dejene F.B, Swart H.C, Ntwaeaborwa O.M, Koao L.F, Motaung T.E, Hato M.J. J. Alloys and Comp. 2016;677:72-79. [15] Zawadzki M, Wrzyszcz J. Mater. Res. Bull. 2000;235:109. [16] Ruifang Z, Lin X, Weifeng Q, Jiansheng C, Biao D, Ligong Z, Hongwei S. J. Mater Sci. 2011;46 (23):7517. [17] Nadia K.Z, Igor B, Andrei K, Dmitri G, José M.F.F. Nanotechnology 2010;21:325707. [18] Ayvacıkli M, Kotan Z, Ekdal E, Karabulut Y, Canimoglu A, Garcia Guinea J, Khatab A, Henini M, Can N. J. Lumin. 2013;144:128–132. [19] de Camargo A.S.S, Nunes L.A.O, Silva J.F, Costa A.C.F.M, Barros B.S, Silva J.E.C, de Sà G.F, Alves Jr S. J. Phys. Condens. Matter. 2010;19:1-7. [20] Murthy K.V.R, Rao C.A, Suresh K, Rao N.V.P, Rao B.S, Walter Ratna Kumar B, Rajasekhar B.N, Rao B.N. Eur. Chem. Techno. J. 2011;13:1-4. [21] Xin M, Hu L.Z, Liu D.P, Yu N.S. Superlattices Microstruct. 2014;74:234-241. [22] Koao L.F, Dejene B.F, Swart H.C. J. Lumin. 2014;147:85-89. [23] Rani G. Powder Technol. 2017;312:354–359. [24] Maphiri V.M, Dejene F.B, Motloung S.V. Res in Phys. 2017;7:3510-3521. [25] Lumbarda V.A. Mech. Mater. 2003;35:53-68. [26] Motloung S.V, Dejene F.B, Swart H.C, Ntwaeaborwa O.M. Ceram. Int. 2015;41:6776-6783. [27] Kroon R.E. S. Afr. J. Sci. 2013;109:(5–6)Art. #a0019.

[28] Chen J.T, Wang J, Zhang F, Zhang G.A, Wu Z.G, Xan P.X. J. Crystal Growth. 2008;310: 2627-2632. 21

[29] Ravikumar B.S, Nagabushana H, Sharma S.C, Nagabushana B.M. Spectro-chim. Acta Part A: Mol. Biomol. Spectrosc. 2014;122:489-498. [30] Motloung S.V, Dejene F.B, Kroon R, Swart H.C, Ntwaeaborwa O.M. Physica B 2015;468469:11-20.

[31] Balda R, Fernández J, Nyein E, Hömmerich U. Opt. Exp. 2006;14:3993-4004. [32] Wang S.F, Sun G.Z, Fang L.M, Lei L, Xiang X, Zu X.T. Sci. Rep. 2015;5:12849.

[33] Kitai A. Luminescent Materials and Applications, John Wiley & Sons, 2008:32. [34] Sernelius B.E, Berggren K.F, Jin Z.C, Hamberg I, Graqvist C.G. Phys. Rev. B: Condens. Matter. Phys. 1988;37:10244-10248. [35] Che Ani N, Sahdan M.Z, Nayan N, Mohd Tawil S.N. Adv. Mat. Res. 2016;1133:424-428. [36] Chanda A, Gupta S, Vasundhara M, Joshi S.R, Mutta G.R, Singh J. RSC Adv. 2017;7:5052750536. [37] Koshy J, George K.C. Int. J. Mat. Phys. 2014;5:35-42.

[38] Liu Y, Luo W, Li R, Zhu H, Chen X. Opt. Exp. 2009;17:9748-9753. [39] Jadwisienczak W.M, Lozykowski H.J, Xu A, Patel B. J. Elec. Mater. 2002;31:776-784. [40] Talewar R.A, Joshi C.P, Mohalir S.V, Opt. Mat. 2016;55:44-48. [41] Tang H, Berger H, Schmid P.E, Levy F, Burri G. Solid State Commun. 1993;87:847850. [42] Caponetti E, Martino D.C, Saladino M.L, Leonelli C, Langmuir. 2007;23:3947-3952

22